Determination of impact of a failure of a component for one or more services

This application is a continuation application claiming priority to Ser. No. 11/014,310, filed Dec. 16, 2004, now U.S. Pat. No. 7,509,518.

FIELD OF THE INVENTION

The invention relates to the field of systems management and in particular to a method and a system for determining the impact of a failure of a component on one or more IT services that the failed component is supporting.

BACKGROUND OF THE INVENTION

Providers of information technology (IT) services are increasingly faced with a demanding, competitive environment in which end users continually expect to be provided with, not only high quality services, but IT services which are available when the user needs them. This has led to providers of IT services ensuring that their IT services are available for use 24 hours a day, 7 days a week and 365 days of the year, in order to meet the demands of their users. If a user is repeatedly faced with IT services that are not available, the end user will soon take their business elsewhere.

The term IT services may comprise many different types of services available to a user for example, on-line banking, an on-line auction system, a payroll management system, a time recording system, a data backup service, or any other type of applications that may be provided by a service provider. Further IT services may comprise core IT services for example data backup, replication of data, producing daily reports and any other type of service that a business needs to be able to carry out its daily activities.

All IT services are supported by information technology systems which may cover the design, development, installation and implementation of IT systems and associated applications, for example, core services such as data backup and replication in order to operate effectively. Each IT system providing an IT service may comprise a plurality of software or hardware components, which may be deployed across a number of different servers and distributed across different geographical regions.

IT service providers offer services to perspective customers that must meet the requirements set down in a service level agreement (SLA). A service level agreement sets down the terms and conditions that shall be met when providing a particular service. For example, the scope of the service to be offered, performance, tracking and reporting, problem management, warranties and remedies, intellectual property rights and confidential information etc.

Often an end user complains of poor customer satisfaction because from the perspective of the end user, the IT service is not available when needed or the IT service gives poor performance at key times of the day. For example, if each employee in a workplace is required to record their worked hours by 3:00 pm each Friday, the time recording system will experience a heavier than normal amount of network traffic logging onto the system. If an IT component fails that is critical to the time recording system then the impact to the business is also critical. Alternatively, from the viewpoint of the service provider the impact of poor performance varies depending on the date and time at which the IT service is running. For example, a user may only log onto a time recording system at 3:00 pm on a Friday afternoon and continually experience poor performance. Therefore from the perspective of the user, the time recording system continually exhibits poor performance. From the service providers viewpoint the impact of poor performance may only be noticed at 3:00 pm on a Friday, because for the rest of the week, the time recording service exhibits a high level of performance due to low user usage.

In order to provide high quality IT services, it is vital for service providers to understand a customer's business and the impact that a failure of a component will have on a particular business requirement. For example, the impact of users not being able to record their work time, as the impact may not only affect the time recording service, but also may have a down stream impact on other IT services that may either be supported by the failed component or the IT service can not start because the service is reliant on the time recording service completing before it can start.

Traditionally users have viewed IT services as a single entity that has a binary state i.e. the IT service is working and available for service or is not available, and therefore the failure of any component that supports that IT service has a similar weight i.e. working or not working. This is often not the case, as a customer's business is made up of many IT services deployed across different components and the IT services may have different availability requirements.

WO 01/79994 describes a business process engine which hosts the execution of electronic business processes and generates event objects for communicating the state of the business process to a business rule engine. The business rule engine hosts the execution of business rule functions triggered dynamically in response to the event object. A business management engine hosts the generation of business reports reflecting system status.

However, this approach has disadvantages because the relationships between a component such as a piece of software or hardware and the IT service that is using that particular component is not identified and therefore a true representation of the availability and criticality of an IT service can not be achieved for any given time period.

OBJECTS AND SUMMARY OF THE INVENTION

Viewed from a first aspect, the present invention provides a method for determining the impact of a failure of a component for one or more services that the component is supporting, the method comprising the steps of: receiving a data feed from a processing node, the data feed containing data indicative of the identity and system status of a component running on the processing node; in response to the identity of the component, first determining if the component supports one or more services; identifying the temporal activity of the one or more services and, in dependence on the determining step, second determining the impact of the system status on the one or more services.

Advantageously, the present invention allows service providers to understand the relationships between a failure of a component and the impact that the failure has on one or more services that the component is supporting. The present invention allows a service provider to not go to the expense of service availability 24 hours a day, 7 days a week, 365 days a year, as the service provider can determine when a user needs a service and when they don't. Further, service providers are able to perform upgrades of their systems without impacting availability of other services.

Preferably the present invention provides a method wherein the identifying step further comprises performing a lookup in a calendar function to identify the temporal activity of the one or more services. The calendar function provides an input mechanism for mapping the temporal activity of each of the services. By analysing and mapping the behaviour of the IT service the daily activity of the IT service can be seen and allows the IT service providers with a projected view into the criticality of their IT services.

Preferably the present invention provides a method, wherein the temporal activity of the one or more services forms one or more rules for use in a rules engine to determine the relationship between the component and the one or more services.

Preferably the present invention provides a method wherein the determined relationships further comprise one or more relationships between services which are dependant on one or more other services.

Preferably the present invention provides a method, wherein the system status of a component identifies whether the component has failed or is active.

Preferably the present invention provides a method wherein in response to identifying that a component has failed, determining the length of time that it will take to rectify the fault and the time in which the component must start to be recovered in order support a scheduled service.

Preferably the present invention provides a method, wherein the result of the determining step produces a list of results identifying each component that has failed and the time period in which each failed component must be recovered.

Viewed from a second aspect, the present invention provides a system for determining the impact of a failure of a component for one or more services that the component is supporting, the system comprising the steps of: means for receiving a data feed from a processing node, the data feed containing data indicative of the identity and system status of a component running on the processing node, in response to the identity of the component, first means for determining if the component supports one or more services; means for identifying the temporal activity of the one or more services and, in dependence on the means for determining step, second means for determining the impact of the system status on the one or more services.

Preferably the present invention provides a system, wherein means for identifying further comprises means for performing a lookup in a calendar function to identify the temporal activity of the one or more services.

Preferably the present invention provides a system, wherein the temporal activity of the one or more services forms means for one or more rules for use in a rules engine to determine the relationship between the component and the one or more supporting services.

Preferably the present invention provides a system wherein the determined relationships further comprise one or more relationships between services which are dependant on one or more other services.

Preferably the present invention provides a system, wherein the system status of a component identifies whether the component has failed or is active.

Preferably the present invention provides a system, wherein in response to identifying that a component has failed, means for determining the length of time that it will take to rectify the fault and the time in which the component must start to be recovered in order support a scheduled service.

Preferably the present invention provides a system, wherein the means for determining time includes means for producing a list of results identifying each component that has failed and the time period in which each failed component must be recovered.

Viewed from a third aspect, the present invention provides a computer program product directly loadable into the internal memory of a digital computer, comprising software code portions for performing, when said product is run on a computer, program steps to carry out the invention.

BRIEF DESCRIPTION OF THE DRAWINGS

Embodiments of the invention are described below in detail, by way of example only, with reference to the accompanying drawings in which:

<figref idref="DRAWINGS">FIG. 1</figref>, is a data processing system in which a preferred embodiment of the invention may be implemented;

<figref idref="DRAWINGS">FIG. 2</figref> illustrates the reprioritization component according to a preferred embodiment of the present invention;

<figref idref="DRAWINGS">FIG. 3</figref> illustrates the output of the reprioritization data into a graphical user interface of the impact assessment module of <figref idref="DRAWINGS">FIG. 2</figref> according to a preferred embodiment of the present invention;

<figref idref="DRAWINGS">FIG. 4</figref> illustrates a component failure and the impacted IT services at a given point in time according to a preferred embodiment of the present invention;

<figref idref="DRAWINGS">FIG. 5</figref> shows the steps that the event reprioritization system performs in order to assess the impact on one or more IT services of a component failure according to a preferred embodiment of the present invention; and

<figref idref="DRAWINGS">FIG. 6</figref> is a timeline of an IT service which is affected by the failure of a component.

DETAILED DESCRIPTION OF THE INVENTION

Where reference is made in any one or more of the accompanying drawings to steps or features which have the same reference numerals, those steps or features have substantially the same function(s) or operation(s) for the purposes of this description.

<figref idref="DRAWINGS">FIG. 1</figref> shows a number of components of a data processing system, in which the invention can be advantageously applied. The data processing system comprises a number of application servers <b>120</b>, <b>150</b>, and <b>160</b> running on processing nodes <b>125</b>, <b>145</b>, <b>140</b> that are connected for communication with one or more user computer's <b>100</b> and <b>110</b>.

The processing nodes <b>125</b>, <b>140</b>, <b>145</b> may be any data processing apparatus which is capable of running an application server, directory server or similar server program. Software-implemented elements of the embodiments described in detail below are not limited to any specific operating system or programming language.

Processing node <b>125</b> comprises a reprioritization component <b>130</b> for interfacing with a monitoring system <b>122</b>. The reprioritization component <b>130</b> receiving real time data feeds from the monitoring system <b>122</b>. The real time data feeds comprising the name of the component, the IP address of the component and the status of the component, for example, the component has ‘failed’, or the component is ‘available’. In response to the information within the real time data feed and further statistical data collected by the reprioritization component <b>130</b> (explained below), an assessment is performed to determine the impact of a component failure on one or more IT services.

In another embodiment a data feed may comprise information regarding the performance of a processing node or component.

For example, the processing node has developed 2 faults in the last 3 days and has experienced down time of 5 hours as a consequence.

Reprioritization component <b>130</b> may be implemented as a computer program using any programming language that is suitable for use within a distributed environment.

Processing nodes <b>125</b>, <b>140</b>, <b>145</b> may run any type of operating system and any type of applications that are able to serve requests from other client or servers located within the data processing system. For example, processing nodes <b>125</b>, <b>140</b>, <b>145</b>, may be running applications that are supporting a customer business, such as a time recording service or a credit checking service.

Processing node <b>125</b> further comprises a non-volatile data store <b>135</b> for storing configuration data about each processing node <b>125</b>, <b>140</b>, <b>145</b>, the temporal nature of each IT service, the results of an impact assessment and the data feeds received from the monitoring system <b>122</b>. The data store <b>135</b> may comprise any suitable archive and retrieval mechanism, for example, a relational database or an XML database. The configuration data may be stored in one single data store or a plurality of data stores depending on the architectural requirements of the data processing system.

Users' computers <b>100</b> and <b>110</b> may comprise an interface <b>105</b> and <b>115</b>, for inputting data into reprioritization component <b>130</b>. The interface may further comprise display means for displaying the output of an impact assessment, a schedular view displaying the activity and criticality of one or more IT services and a view of the current system state of one or more IT Services.

User's computer <b>100</b>, <b>110</b> is not limited to a particular type of data processing apparatus, and may be a conventional desktop or lap-top personal computer, a personal digital assistant (PDA) or other specialized data processing device which is capable of running an application program. The user's computer may connect to a network of data processing systems via wireless or hard-wired connections. Communication between user's computers <b>100</b>, <b>110</b> may be carried out over a network such as a local area network, a wide area network or across the Internet. Each of user's computers <b>100</b>, <b>110</b> and processing nodes <b>125</b>, <b>140</b>, <b>145</b> may be remote from each other within the network.

In an example embodiment, processing nodes <b>145</b> and <b>140</b> are providing a plurality of IT services. For example, an application server may be providing a credit checking service for a particular customer and a time recording process for another customer. Alternatively processing nodes <b>145</b> and <b>140</b> may be providing core IT services such as producing daily reports or a replication service.

An IT service may be distributed across a plurality of processing nodes. For example, a distributed database serving data to multiple geographically dispersed web sites. For each of these examples there may be common hardware or software components which are supporting the credit checking service, the time recording process or the geographically dispersed web sites. Therefore, it can be seen that there is not a one to one relationship between a hardware or software component and an IT service, but in fact many components may support many IT services i.e. a many to many relationship.

The impact of a component failure may depend not only on which component has failed and which IT service is impacted but also the implementation of the IT service. For example, a particular component/service combination deployed by a particular company when interrupted by a component failure may result in a different impact when deployed by different company. For example, when deploying a supply chain management service, one of the failed components used to deploy this service may be a DB<b>2</b> database. For company ‘A’ the predicted recovery time for the component failure may be 4 hours, while for company ‘B’, the predicated recovery time for the same component may by 8 hours. The predicted recovery time for company ‘B’ may be longer due to a high volume of data input into the service.

In order to provide users with a particular level of service, it is an important requirement for any system operations manager to understand the relationships between a component and the IT services that it may support. This enables component upgrades to be performed, system outage time to be minimized and IT service problems to be reported and rectified in the quickest possible time.

Referring to <figref idref="DRAWINGS">FIG. 2</figref>, the reprioritization component comprises the following components, publisher component <b>200</b> for collating and displaying information gathered by the system, authentication component <b>210</b>, for authenticating one or more data feeds received from the monitoring system and impact assessment component <b>220</b> for determining the impact on a component of disruption from its normal operation on one or more IT services. Each one of these components will now be explained.

The publisher component <b>200</b> provides an interface for receiving one or more data inputs from user's computer <b>100</b>, <b>110</b> and displaying one or more outputs of reprioritization component <b>130</b>.

In order for the system to be able to perform an impact assessment, the system requires data from a number of sources. One particular source is data regarding the temporal nature of an IT service. A service exhibits different behaviour at different times of the day, month and year, depending on the requirements of the users using the services. For example, using an example of a time recording service, the users of the time recording service may record their work hours at 15:00 on a Friday afternoon. This requirement is dictated by the users of the time recording service. Similarly, towards the end of a financial year, an expense reporting service may experience heavier use by users than at other times of the year. Therefore, the temporal activity of a service may comprise the days and times of when the service should be running and when the service must run because of the need to have the service working when the users need it most. Further, the criticality of the service (which describes the level of importance attributed to the service being available and is defined by an amount of time or permissible disruption to normal operation) may vary for indirect reasons. For example, a customer relationship management service for a sales force may become more critical at the time a new product is launched.

The temporal activity of an IT service is mapped into a scheduler function. The scheduler function comprises the ability for the temporal activity of an IT service to be mapped against minutes, hours, days, months and years.

Referring to <figref idref="DRAWINGS">FIG. 3</figref> which depicts a table, the mapping of a number of IT services can be seen within the calendar function. The activity of the IT service is broken down into the temporal activity of IT service <b>310</b> over each day of the week <b>300</b>. For example, IT service contract management <b>330</b> must run Mondays to Friday between the hours of 04:00 to 20:00 hours. In contrast, IT service daily reports <b>315</b> must run Monday to Friday but only between the hours of 01:00 and 7:00 hours.

Each IT service mapped into the calendar function may further be categorized by its level of criticality. For example, contract management IT service <b>330</b> may be further categorised into a particular color band which signifies in the event of a component failure that it is ‘important’ <b>325</b> that the IT service runs at a particular time of the day.

Similarly if the IT service is categorised as ‘critical’ <b>320</b> then a further color banding may be used to signify this status. Each time, impact assessor <b>220</b> calculates the impact of a failure of a component on the one or more IT services that it supports, the criticality of each IT service may be dynamically viewed over a period of time. The calendar function is therefore a store for the changing criticality of an IT service which is dynamically updated by the reprioritization component <b>120</b>.

The calendar function enables time zones of different countries to be managed. For example, a help desk agent working in the UK, who is talking to a user in Germany and whose system, is being maintained in the US. Each different user involved in this example will require the same information, but adjusted to their own time zone.

The calendar function further provides a dynamic view of one or more IT services over a period of time, displaying the criticality of an IT service as it is affected by one or more component failures. For example, a component may fail that affects a time recording process. The component has failed at 14:00. A failure of the component at 14:00 is deemed not to affect the time recording process and is displayed in the scheduler as an amber failure. It is critical that the time recording process is available for use at 16:00, therefore as the current time moves closer to 16:00 and the component failure has not been fixed the IT service may be displayed as red, signifying a critical failure. If the failure of a component is not deemed to affect an IT service at a particular point in time, a calculation is performed to determine to identify when the IT service would be considered at risk.

The temporal activity of each of the IT services that a component is supporting is stored in data store <b>135</b>. For example, table 1 illustrates the temporal nature of a time recording process.

<tables id="TABLE-US-00001" num="00001"><table frame="none" colsep="0" rowsep="0"><tgroup align="left" colsep="0" rowsep="0" cols="3"><colspec colname="offset" colwidth="63pt" align="left" /><colspec colname="1" colwidth="140pt" align="center" /><colspec colname="2" colwidth="14pt" align="left" /><thead><row><entry /><entry namest="offset" nameend="2" rowsep="1">TABLE 1</entry></row></thead><tbody valign="top"><row><entry /><entry namest="offset" nameend="2" align="center" rowsep="1" /></row><row><entry /><entry>Service</entry><entry /></row></tbody></tgroup><tgroup align="left" colsep="0" rowsep="0" cols="4"><colspec colname="offset" colwidth="63pt" align="left" /><colspec colname="1" colwidth="56pt" align="left" /><colspec colname="2" colwidth="56pt" align="left" /><colspec colname="3" colwidth="42pt" align="left" /><tbody valign="top"><row><entry /><entry>Should run</entry><entry>Must run</entry><entry>Critical</entry></row><row><entry /><entry namest="offset" nameend="3" align="center" rowsep="1" /></row></tbody></tgroup><tgroup align="left" colsep="0" rowsep="0" cols="5"><colspec colname="offset" colwidth="14pt" align="left" /><colspec colname="1" colwidth="49pt" align="left" /><colspec colname="2" colwidth="56pt" align="left" /><colspec colname="3" colwidth="56pt" align="left" /><colspec colname="4" colwidth="42pt" align="left" /><tbody valign="top"><row><entry /><entry>Time</entry><entry>Monday 6:00</entry><entry>Friday 8:00</entry><entry>Friday</entry></row><row><entry /><entry>recording</entry><entry>to Thursday</entry><entry>to 12:00</entry><entry>12:00 to</entry></row><row><entry /><entry /><entry>17:00</entry><entry /><entry>17:00</entry></row><row><entry /><entry namest="offset" nameend="4" align="center" rowsep="1" /></row></tbody></tgroup></table></tables>

The temporal activity stored for each IT service, therefore forms one or more rules for each IT service, which can be propagated down through the use of a rules engine, to all the processing nodes within the data processing system. For example, the day is Monday and the time is 12:00, according to Table 1, this means that the time recording process ‘should run’. Using a further example, the day is Friday and the time is 13:00, therefore according to Table 1, the time recording service is classified ‘as critical’. Each of these rules can be defined in a rules engine within impact assessor <b>220</b>. As each of the data feeds are received by impact assessor <b>220</b>, the rules engine can define the criticality of each of the IT services supported by a component that has failed or developed a fault.

Once the temporal nature of an IT service has been defined, the results from the rules engine are mapped to the components that support the one or more IT services.

For example,

<ul id="ul0001" list-style="none"><li id="ul0001-0001" num="0063">Payments to Canada—&lt;needs hardware component X, software component Y—to run&gt;</li><li id="ul0001-0002" num="0064">Payments to USA—&lt;needs hardware component X, software component Y—to run&gt;</li></ul>

The mapping may be carried out by populating an XML schema. The XML schema may comprise the following IT service-&gt;components-&gt;nodes. Therefore failure of nodes and component may be monitored and propagated back up to the IT service that they are supporting.

Referring to <figref idref="DRAWINGS">FIG. 4</figref>, the impact of a component failure can be seen in relation to a number of IT services at a given point in time <b>400</b> i.e. the illustrative time in <figref idref="DRAWINGS">FIG. 4</figref>, being 10:30. Each IT service is categorised into ‘must run’ <b>410</b>, ‘should run’ <b>401</b> and ‘not needed’ <b>450</b>. Once a data feed has been received of a component fault of failure that impacts its normal operation, the component that has failed or developed a fault is displayed in the ‘failed’ column <b>405</b>.

Therefore it can be seen that the USA hardware node <b>415</b> and USA package <b>415</b> has failed. This in turn impacts <b>470</b>, the settlement service <b>420</b>, and the USA payment service <b>420</b>. The most important of these services, is the European payments which is running <b>430</b> but as of yet not impacted.

Equally, it can be seen that at 10:30 <b>400</b> the services that should run <b>401</b> is the USA payment service <b>440</b>. This service is also impacted by the failure of the USA hardware node <b>435</b>. Further at 10:30 the Canada payment service <b>445</b> is running and may be impacted by the USA hardware node failure.

Further, in the ‘not needed’ row <b>450</b>, it may that the USA Hardware node <b>455</b> is categorised as ‘not needed’. Therefore if the ‘end of day reporting’ <b>460</b> is the only IT service that is impacted or the ‘Japan payment’ service <b>465</b> is running, then it is possible for a systems operation manager to perform a system outage operation. Hence it is possible to understand which components are supporting one or more IT services and in the event of a component failure, take the most appropriate remedial action.

By understanding the relationships between a failure of a component and the one or more IT services that it supports the effect of the impact may be understood. For example if service X runs at 10:00 and takes 2 hours to complete, it may be that services X and Y can only run after service X had completed. Therefore the ‘knock on effect’ of failure or delayed running may be understood.

Information regarding each component residing on a processing node may be gathered by a configuration management system. As is known in the art, configuration management identifies how various IT components (e.g., hardware, software, service-level agreements, documentation, databases, policies and procedures) are linked together within an IT environment, for example, a workstation ‘X’ has an operating system installed of a particular time and a number of other software applications.

The configuration management system allows the tracking of a processing node's <b>135</b> hardware and software configuration including physical and logical network information and administrative information such as personnel information and location information, etc. To enable configuration data to be extracted from each processing node within a data processing network, as is known in the art, a configuration agent is installed on each processing node. The processing node may be a client computer or a server computer running a variety of software applications.

The configuration agent extracts the configuration data in the form or a flat file. This file is sent to an importer agent (not shown) installed on a server. The importer agent extracts the configuration data from the flat file and writes the configuration data to a data store. SQL queries can be performed on the configuration data, stored in a data store (not shown).

The configuration data gathered from each processing node may comprise detailed listings of the hardware and software installed on each processing node. An example of a configuration listing for a processing node is as follows: <ul id="ul0002" list-style="none"><li id="ul0002-0001" num="0075">Server serial number: 78XXB31</li><li id="ul0002-0002" num="0076">Operating system (OS): AIX</li><li id="ul0002-0003" num="0077">OS version number: 5</li><li id="ul0002-0004" num="0078">Applications: Transaction verification V2.6 <ul id="ul0003" list-style="none"><li id="ul0003-0001" num="0079">Risk engine—high risk V8.91</li></ul></li></ul>

The configuration data for each processing node is stored in an XML schema. The configuration data defined within the XML schema listing each hardware and software component of a processing node. The XML schema is modified to further comprise the name of the IT service that a component is supporting, and the process start time, and when the IT service should be running, and when it is critical that it is running.

The publisher component is also responsible for providing an input and output interface for simulation of the impact that a component failure has on an IT service. The user is able to enter commands into the interface to simulate certain scenarios. For example, it may be necessary to upgrade a particular hardware component and therefore, it is vital to understand which IT services the hardware component is supporting. The presentation component is able to display the calculated impact assessment to the user.

Turning now to monitoring component <b>122</b>, monitoring component <b>122</b> receives one or more data feeds from one or more processing nodes. The monitoring component may ‘ping’ the one or more processing nodes every predetermined number of seconds to determine whether a component is running or has developed a fault. A ‘ping’ requires the requesting processing node to request the IP address of the receiving processing node. For example,

Ping www.ibm.com may return the IP address 127.0.0.1 and the following information.

<ul id="ul0004" list-style="none"><li id="ul0004-0001" num="0000"><ul id="ul0005" list-style="none"><li id="ul0005-0001" num="0083">Pinging www.ibm.com [129.42.18.99] with 32 bytes of data:</li><li id="ul0005-0002" num="0084">Reply from 129.42.18.99: bytes=32 time&lt;1 ms TTL=128</li><li id="ul0005-0003" num="0085">Reply from 129.42.18.99: bytes=32 time&lt;1 ms TTL=128</li><li id="ul0005-0004" num="0086">Reply from 129.42.18.99: bytes=32 time&lt;1 ms TTL=128</li><li id="ul0005-0005" num="0087">Reply from 129.42.18.99: bytes=32 time&lt;1 ms TTL=128</li><li id="ul0005-0006" num="0088">Ping statistics for 129.42.18.99:</li><li id="ul0005-0007" num="0089">Packets: Sent=4, Received=4, Lost=0 (0% loss),</li><li id="ul0005-0008" num="0090">Approximate round trip times in milli-seconds:</li></ul></li></ul>

Minimum=0 ms, Maximum=0 ms, Average=0 ms

The above information indicates that the processing node or component is active. If the processing node or component was not active the following may occur <ul id="ul0006" list-style="none"><li id="ul0006-0001" num="0000"><ul id="ul0007" list-style="none"><li id="ul0007-0001" num="0093">Reply from 129.42.18.99: bytes=32 time&lt;1 ms TTL=128</li><li id="ul0007-0002" num="0094">Reply from 129.42.18.99: bytes=32 time&lt;1 ms TTL=128</li><li id="ul0007-0003" num="0095">Ping statistics for 129.42.18.99:</li><li id="ul0007-0004" num="0096">Packets: Sent=4, Received=0, Lost=4 (100% loss),</li><li id="ul0007-0005" num="0097">Approximate round trip times in milli-seconds:</li></ul></li></ul>

Minimum=0 ms, Maximum=0 ms, Average=0 ms

In the above example, it can be seen that the number of packets sent were 4 and that the receiving machine did not receive the 4 packets and hence all 4 packets were lost, indicating a failure of the processing node or component.

Alternatively processing nodes, hardware or software running on a processing node that does not have an IP address, a VTAM command may be used to return the status of a component. Other means for obtaining information concerning the status of a component comprise parsing application logs and operating system logs. In another embodiment a publish subscribe mechanism may be used to publish specific information that is required from one or more processing nodes.

The data feeds comprise information regarding the processing nodes host name, the processing nodes IP address, the name of the component that is supporting the IT service and information regarding the state of the component, for example, state=running, or state=failed. The monitoring component routes each of the received data feeds onto the authentication component.

Turning now to authentication component <b>210</b>, authentication component <b>210</b> ensures that a data feed which is received from the monitoring component is a valid data feed and that a user who is logged on to the impact assessor is a valid user. The validation of a data feed requires the authentication component to perform a look up in a data store to locate if the IP address of a processing node is a registered IP address. The authentication component further detects that the data feed carries an authentic user name and password otherwise the data feed with be returned to the requesting client device <b>100</b>, <b>110</b> or processing node <b>140</b>, <b>145</b>.

Once the data feed has been authenticated it is sent to impact assessor component <b>220</b>. If it is identified in the data feed that a component has developed a fault, the impact assessor component collates for each component identified in the data feed information regarding the one or more IT services that it may be supporting. Table 2 is an example of the type of data that may be retrieved from data store <b>135</b>.

<tables id="TABLE-US-00002" num="00002"><table frame="none" colsep="0" rowsep="0"><tgroup align="left" colsep="0" rowsep="0" cols="1"><colspec colname="1" colwidth="217pt" align="center" /><thead><row><entry namest="1" nameend="1" rowsep="1">TABLE 2</entry></row></thead><tbody valign="top"><row><entry namest="1" nameend="1" align="center" rowsep="1" /></row><row><entry>EMEA Order Fulfillment,</entry></row><row><entry>Ship Confirmation (a), BRT: 8.00 hrs, Criticality: Important</entry></row><row><entry>Start Date: 00:00 May 12, 2003, End Date: 23:59 May 12, 2003,</entry></row><row><entry>impact = Data</entry></row></tbody></tgroup><tgroup align="left" colsep="0" rowsep="0" cols="5"><colspec colname="offset" colwidth="21pt" align="left" /><colspec colname="1" colwidth="56pt" align="left" /><colspec colname="2" colwidth="56pt" align="left" /><colspec colname="3" colwidth="42pt" align="left" /><colspec colname="4" colwidth="42pt" align="left" /><tbody valign="top"><row><entry /><entry>Component</entry><entry>Location</entry><entry>Host</entry><entry>Type</entry></row><row><entry /><entry namest="offset" nameend="4" align="center" rowsep="1" /></row></tbody></tgroup><tgroup align="left" colsep="0" rowsep="0" cols="3"><colspec colname="offset" colwidth="21pt" align="left" /><colspec colname="1" colwidth="154pt" align="left" /><colspec colname="2" colwidth="42pt" align="left" /><tbody valign="top"><row><entry /><entry>H5Router</entry><entry>Fault</entry></row><row><entry /><entry namest="offset" nameend="2" align="center" rowsep="1" /></row></tbody></tgroup></table></tables>

Using the data within Table 2, a data feed may be received from the authentication component indicating that the H5Router has developed a fault. A look up is performed in data store <b>135</b> to locate the IT service that the component is supporting. In this example, the IT service is the EMEA Order Fulfilment. Further information is retrieved concerning the business recovery time, the criticality, the start date and start time, the end date and end time (Fatal, Data, Performance) of the running of the EMEA Order Fulfilment service.

Further information may be retrieved from data store <b>135</b> to indicate when key milestones are met by an IT service. For example starting bulk payment at 18:00 hours and completed bulk payment at 19:01 hours. Thus allowing the monitoring of an IT services success due to completion times.

Once the above data has been gathered, impact assessor component <b>220</b> begins to calculate the latest time in which a component must be replaced, repaired or updated, in order to minimize the impact on one or more IT services.

Referring to <figref idref="DRAWINGS">FIG. 5</figref>, the steps the impact assessor performs in order to create an impact assessment are explained:

At step <b>500</b> a data feed is received by the authentication component from a processing node. The authentication component parses the data feed to determine whether the data feed is a valid data feed at step <b>505</b>.

Alternatively, at step <b>500</b>, the user may issue instructions from user's computer <b>110</b>, <b>110</b> requesting a simulation to be carried out on the impact of an IT service on one or more components failing. The authentication component receives the instructions in the form of a data feed and authenticates the data feed to ensure that the data feed is a valid data feed at step <b>510</b>.

Once the data feed has been authenticated, a determination step is performed at step <b>520</b>, to establish whether the component is supporting an IT service. A query is performed in the data store at step <b>520</b> to determine which IT service the component is supporting <b>525</b>. A list is retrieved from the configuration database detailing each of the IT services that the component is supporting (please refer to Table 2).

Returning to step <b>515</b>, if the determining step is negative, a response is sent back to the user, informing the user <b>570</b> that the component is not supporting one or more IT services. In this instance, if the component needed to be upgraded, the user would be able to perform the upgrade knowing that the upgrade would not impact any IT services.

Once it has been determined that the component is supporting one or more IT services, the data feed is further authenticated at step <b>535</b>, a query is performed at step <b>540</b> in the data store <b>540</b> to determine if the IT service is active at the present time <b>545</b>.

Alternatively, if the query finds that the IT service is not active <b>550</b>, then a response informing the user that the IT service is not active is sent back to the user at step <b>570</b>.

At step <b>555</b> the data feed is further authenticated and the impact assessor determines the severity of the failure of the component on the one or more identified IT services at step <b>560</b>.

Once the impact assessor determines the impact of the failure of a component on one or more IT services, the response is built <b>570</b> and displayed in the schedular function <b>200</b> at step <b>575</b>.

Referring to <figref idref="DRAWINGS">FIG. 6</figref>, a timeline illustrates the steps of determining the severity of the impact and calculating the overall recovery time for a component failure. At step <b>565</b>, the impact assessor analyzes the gathered information obtained by steps <b>505</b> to <b>555</b> of <figref idref="DRAWINGS">FIG. 4</figref>.

The impact assessor begins by establishing the time (t<b>0</b>) <b>600</b> at which the data feed was received by the impact assessor, as this is the time that it is determined that the impact has failed. The impact assessor parses the data feed to locate a time stamp indicating the time the data feed was received by the authentication component and the time in which the component has failed.

There may be a time lapse between the time at which the component failed and the time in which a system operator is alerted to the failure of a component. Therefore a further time stamp is taken at the time a system operator acknowledges the component failure <b>605</b>. The acknowledgement may take the form of an input button that signifies that the system operator has started to process the component failure.

Next, impact assessor <b>220</b> determines the latest time in which the failed component may be recovered (business recovery time (BRT)) <b>625</b>. The BRT <b>625</b> indicates ‘allowable interruption time’ to an operational service from a single event. A query is performed within data store <b>135</b> to locate when the component must be running to support one or more IT services. For example, it may be recorded in data store <b>135</b> that a time recording process must be available for use at 15:00 pm on Friday (T<b>2</b>) <b>620</b>.

A further lookup is performed in data store <b>135</b> to determine how long it will take to bring ‘back on line’ the failed component (predicted overall recovery time) <b>615</b>. For example, if the component needed to be replaced, it could take 1 hour to carry out the upgrade and bring the processing node back on line.

Based on the variables defined above the impact assessor can calculate the business recovery time for any given component for one or more component failures. For example, using the example above, the latest time <b>610</b> in which a component must be recovered=T<b>2</b>−Predicted overall recovery time, i.e. the latest time a component must be recovered=15:00−60 mins=14:00.

A further calculation is performed to determine the latest possible time <b>630</b> that the service must be available i.e. the slack time in which to start repairs of the components. For example, <br /><i>T</i>3=<i>T</i>2+BRT latest possible time the service must be available.<br /> Again using the previous example, T<b>3</b>=14:00 (IT service start time)+120 mins=16:00 hours.

The above calculations are performed for each component failure in real time as the component failures are notified to the impact assessor.

The results of each of the above calculations produce a list of components and the time in which each of the components must be started. The list may be sorted in many ways, but in a preferred embodiment the list is sorted by recovery sequence.

The results of the calculations may result in groups of components for which recovery must be started at the same time. Within these groups, an order of recovery is determined. In a preferred embodiment the order of recovery is determined, for example, by the type of impact of a failure, for example, fatal, critical or available.

Within each of these categories, the order may be determined by whether or not the failure will have a downstream impact on other IT services or other components, or an indirect impact on other IT services. Thus, a failure of a component with a downstream dependency will take precedence over a failure of a component with no downstream dependency.

For example, the table below illustrates a number of impacts and their associated dependencies.

<tables id="TABLE-US-00003" num="00003"><table frame="none" colsep="0" rowsep="0"><tgroup align="left" colsep="0" rowsep="0" cols="4"><colspec colname="offset" colwidth="21pt" align="left" /><colspec colname="1" colwidth="77pt" align="left" /><colspec colname="2" colwidth="56pt" align="left" /><colspec colname="3" colwidth="63pt" align="left" /><thead><row><entry /><entry namest="offset" nameend="3" align="center" rowsep="1" /></row><row><entry /><entry>Component</entry><entry>Impact</entry><entry>Dependency</entry></row><row><entry /><entry namest="offset" nameend="3" align="center" rowsep="1" /></row></thead><tbody valign="top"><row><entry /><entry>Router</entry><entry>Fatal</entry><entry>Down stream</entry></row><row><entry /><entry>Disk drive</entry><entry>Fatal</entry><entry>No</entry></row><row><entry /><entry>Network adapter</entry><entry>Critical</entry><entry>Down stream</entry></row><row><entry /><entry>Memory failure</entry><entry>Critical</entry><entry>No</entry></row><row><entry /><entry>Router</entry><entry>Available</entry><entry>Downstream</entry></row><row><entry /><entry>Processor</entry><entry>Available</entry><entry>No</entry></row><row><entry /><entry namest="offset" nameend="3" align="center" rowsep="1" /></row></tbody></tgroup></table></tables>

There may also be a “recovery dependency hierarchy” for the affected components which will be factored into the sequence, within the above impacts. Recovery dependent components inherit the priority of their ‘parents’, and are placed one above the ‘parent’ in the sequenced list.

The output from the prioritization may be a display illustrating all the components that need to be recovered, along with the component's dependency information. A preferred sequence of display is in order of latest recovery time.

Logic is applied to sequence the recovery actions, within the overall prioritization given, taking into account tacit knowledge, availability of resources and (optionally) feedback from the affected customers, to determine the actual recovery plan.